If we were to look at the future of healthcare purely on buzzwords, here's what's in vogue right now: personalised, predictive, and digital. Add AI to the mix, and you've got the potential to transform our approach to how we understand, monitor, and treat illness and deliver medical services.
Oxford Immune Algorithmics (OIA) is a UK healthcare startup spun out of the University of Oxford.
The company is on a mission to advance towards predictive precision medicine.
It has developed a platform called Algocyte that integrates information from blood tests with other patient data and academic research into medical conditions to analyse a patient's health.
Algocyte remotely monitors patients and offers healthcare providers early warnings about potential issues.
I recently spoke to Dr. Hector Zenil, founder of Oxford Immune Algorithmics, at IFA, Germany's consumer electronics trade show.
"Algocyte learns from the user and builds a haematological digital twin over time, a version of the individual's baseline health data against which to track deviations from their personal optimal health remotely and in a timely fashion."
Besides the digital platform, which is currently able to deliver, process and ingest data from its own lab partners, OIA is developing handheld devices that allow physicians to regularly monitor their patients' condition through at-home blood tests.
Each test on the Algocyte device will use a different cartridge, each requiring a small blood sample. The device will allow people to get results on the spot, starting with a complete blood count test, the most commonly conducted medical test.
By combining medical data and both predictive and generative AI, OIA wants to develop a generalisation of AI capable of reasoning like scientists that tracks a patient's data regularly to understand their personal health dynamics.
The importance of immune cell changes in health
The company was founded in 2018 and is underpinned by an increasing recognition that the immune system plays a critical role in the pathogenesis of most, if not all, diseases. Analysis of the immune system using AI will enable more targeted and personalised treatments.
Understanding immune cell changes also helps maintain and prolong good health over time. "We monitor the monitors to help people live healthier longer", said Dr. Zenil.
OIA contends that there is a general lack of understanding of the changes in relating different immune cell types (counts/proportions and/or markers) and their associated change over time as they transit moving from healthy to unhealthy states – such as individual diseases, or groups of diseases (say inflammatory ones), or to ageing.
Algocyte endeavours to match the immune system's complexity in understanding how individuals transition from health to disease.
A blood test is one of the first lines of investigation to assess the immune system's performance.
Traditionally, medical tests work with average population reference values, a sort of systematic human bias.
For example, the total number of white blood cells within a range of 4 – 11 million cells per millilitre is considered normal and anything outside would be in the abnormal category, dramatically simplifying into a binary decision the clinical and biological state of a person. If you fall within that range, you are considered normal.
However, clinical research has demonstrated that variations in immune-related cells within the 'normal' range can have important health implications, especially those that are rapid.
In one study, the total white blood cell (WBC) count in 'healthy' individuals was found to be predictive of risk of short-term and long-term death, even when factoring in controlled risk factors like age, smoking, diabetes, high blood pressure, ethnicity and blood cholesterol levels.
Another study found that those with a high 'normal' WBC were almost three times more likely to die within six months than those with a lower 'normal' WBC.
From the lab to the customer
I was curious to learn how OIA would gain traction with doctors who are often bombarded with potential new subscriber digital platforms, SaaS, and various product offerings.
According to Dr. Zenil, efficiency is key. He explained:
"Our solution can, for example, reduce the 25 to 50 percent of cancer patients that do not die because of their tumour growth but because of an infection whose symptoms were confused with the adverse effects of chemotherapy.
The solution is to monitor these patients more closely and know what are the subtle differences in the signatures of sepsis versus the chemo. This will be a new standard of care".
Dr. Zenil predicts that change will most likely come from three different forces:
- The healthy consumer dissatisfied with their current standard of care, huge waiting lists and dangerous backlogs.
- Public and private insurers that find reactive sick care is many times more expensive than proactive health care.
- Life science companies, wanting to push medical services to people's homes with the right incentives will help them with current issues related to adherence and penetration.
Dr. Zenil explained:
"For example, there are fantastic viral drugs in the market, but they are heavily under-prescribed in favour of antibiotics because bacterial infections are more common, and antibiotics have dominated the market to the detriment of large numbers of patients suffering from viral infections with virtually no access to proper diagnosis."
OIA works very closely with the NHS.
In a large NHS GP practice survey by the company, doctors and nurses said they could save up to 3 hours using an automation tool for patient triaging and patient analysis like Algocyte, even if it meant switching between different SaaS solutions or platforms.
As a tech journalist, I hear about startups and blood testing, and of course, my brain remembers Theranos. But Dr. Zenil told me the company has found way less backlash from patients than investors who became highly risk-averse after Theranos.
"Patients usually rely, for better or worse, on regulatory approvals. We are focused on validating our technology in the medical space and hold ourselves to the highest standards having already achieved ISO certification and early CE and UKCA accreditation for our software and algorithms.
Our partners and stakeholders rely on the experience and credentials of our highly qualified team, who are world experts in their fields."
Hardware is hard
Ok, I'm never going to stop saying "hardware is hard". Or to put it differently, highly ambitious, especially when you are offering a consumer device.
Dr. Zenil explained:
"There is no other device that performs sophisticated blood testing other than rather simplistic ones based on single biochemistry markers such as glucose metres.
By tackling a much more sophisticated medical test, we know we can rapidly expand to the easier ones hence covering all key panels to better assess a patient or user health.
But we also integrate with other devices and data streams. Our AI platform is agnostic and works with file standards like HL7 widely used by laboratories and other devices."
OIA's solution will be launched in stages as the company strives to keep innovating. The medical instrument will come next year, but it will launch most of its technology based on the AI platform by the next quarter.
Lead image via Oxford Immune Algorithmics. Photo: Uncredited